Decision Trees, Survival Trees, and Random Forest: Practical Examples with R Programming
In this session of the BTEP Coding Club, Brian Luke, PhD, Senior Principal Computational Scientist with the Advanced Biomedical Computational Science (ABCS) group, demonstrated the use of R programming to perform decision tree analysis, survival tree analysis, and random forest.
The session covered the following:
-
Decision Tree Analysis
The decision tree analysis used the “kyphosis” dataset to predict the absence or presence of kyphosis (a type of deformation) following corrective spinal surgery.
-
Survival Tree Analysis
The survival tree analysis used the recurrence-free survival time from a prospective randomized clinical trial conducted by the German Breast Cancer Study Group.
-
Random Forest
Random forest was applied to the German Credit Data set to determine whether they should or should not receive a loan of a given amount.
R Script and Related Materials
R script
Access the R script used in this tutorial here.
Related Materials
For a more detailed theoretical background on these topics, check out this related presentation ("Decision Trees, Survival Trees, and Random Forest") also by Brian Luke, Ph.D.